how to reduce cost per SQL on google ads for SaaS

here's a sentence i hear constantly from SaaS founders: "our google ads are generating leads, but none of them are closing."

sometimes that's a sales problem. more often it's a paid media problem — specifically, that the account is optimized for volume instead of quality, and google has been trained to find whoever is most likely to fill out a form, regardless of whether they could ever buy.

getting google ads to drive actual pipeline is a different job than getting it to drive leads. here's how to do it.

the real problem

google's smart bidding is genuinely impressive. it will find conversions at a rate that will make your CPL look great in a dashboard. the catch is that it optimizes for whatever you tell it a conversion is — and most SaaS accounts tell it that a form fill is a conversion.

so google finds form fillers. efficient, cheap, plentiful. and mostly useless if your actual goal is qualified pipeline.

the algorithm is not your enemy. it's doing exactly what you asked it to do. the problem is what you asked.

the fix starts with giving google better signal about what a good conversion actually looks like.

account structure that actually matters

most SaaS google ads accounts are structured around what's easy to manage, not what drives performance. you end up with broad campaigns, mixed intent keywords, and everything pointing to the same landing page.

what actually works:

• separate campaigns by intent tier. bottom-of-funnel keywords ("salesforce alternative," "hubspot competitor," "[your category] software") get their own campaign with tighter bids and dedicated landing pages. top-of-funnel gets its own budget and expectations set accordingly
• brand gets its own campaign. always. mixing brand and non-brand muddies performance data and usually results in non-brand keywords subsidizing brand CPCs
• competitor keywords get their own campaign. they behave differently, convert differently, and need different creative and landing pages

keywords: go narrower than you think

broad match has improved a lot. it has also caused a lot of accounts to hemorrhage budget on searches that have nothing to do with what they sell.

for bottom-of-funnel campaigns, phrase and exact match give you more control and more predictable intent. check your search terms report weekly. if you're seeing searches that would never convert, add them as negatives immediately. don't wait for the monthly review.

for SaaS specifically, some useful negative categories to build out: job seekers (add "jobs," "salary," "career," "how to become"), students and researchers ("what is," "definition," "wikipedia"), and anyone looking for free stuff ("free," "open source," "crack," "torrent") unless you have a free tier and actually want them.

bidding for pipeline, not leads

if you have enough data — usually 30+ conversions per month — target CPA or target ROAS bidding outperforms manual. the question is what conversion action you're optimizing toward.

ideal setup:

• import your CRM data so google can see which form fills turned into opportunities or closed deals
• set your primary conversion action as MQL or SQL, not raw form fill
• if you don't have enough data for smart bidding on downstream events, use a micro-conversion (demo booked, not just form submitted) as a proxy while you build data

this takes longer to optimize. it's worth it. an account optimizing toward SQLs will find SQLs. an account optimizing toward leads will find leads.

tracking: if you don't have this set up, nothing else matters

i cannot stress this enough. before you touch a bid strategy, an audience, a keyword, or a creative — make sure your conversion tracking is correct.

specifically:

• conversions should only fire once per lead, not on every page visit or thank-you page reload
• if you're using google tag manager, QA every conversion action in debug mode before you trust it
• import offline conversions from your CRM. if you're not doing this, you're flying blind on what's actually driving revenue
• check that auto-tagging is on and that gclid is being captured in your CRM

i've audited accounts where 40% of reported conversions were duplicates. the account looked like it was printing leads. it wasn't.

honest take

reducing cost per SQL is mostly about giving google better information and being more disciplined about what you're asking it to do. it's not glamorous. there's no secret bidding strategy or magic keyword.

the accounts that drive real pipeline from google ads are the ones where someone is actually paying attention to search terms, feeding CRM data back in, and not celebrating a low CPL without asking what happened to those leads afterward.

if your google ads are generating activity but not pipeline, let's look at it together. usually takes about 20 minutes to figure out where the problem is.